Thursday, March 31, 2005

It's worse than I thought, Only the top 20% are worse off!

In my previous post, I mentioned that perhaps 60% of the population were better off according to the IFS figures. Turns out that I was wrong. It's more like 80%.

Got that? The headlines you read in the paper yesterday about average income dropping were driven by a small decline in the take home pay of the top 20% of Households.

Don't believe me? OK, here is the IFS press release. At the bottom you will find a table listing Average and Median income changes by year.

Directly above the table is a weblink to the DWP website, analysing Households with Below Average income. If you go to that site you eventually get to these statistics, which appear to present the same IFS data- (the last two columns are the same as the first two columns in the IFS table). Thing is, they also present the median income of each 10% of the population. (the first.. umm ten columns. Obviously.)

Now what to the decile median incomes teach us (apart from the fact that phrases like median decile kills reader interest stone dead)? Why, that Income went up for the first eight deciles. Gosh. So the bottom 80% of the population were better off, even when average income went down? Why yes indeed, it does, pretty much*.

So let's take that from the top folks. The data that you read in the paper this morning telling you that average incomes had fallen in big glaring headlines, actually shows that at the very least 75% of households were better off than the year before.

That's according to one bloke with no statistical or economic training at all, who only had his first thought about incomes and earnings data while reading his copy of the Times on the bus to work. Care to prove me wrong?

Of course, If I'm wrong, I'll apologise with exactly the same vigour that I critique. Anyone reckon the Mail, the Times and the Standard will do the same?

*OK if you want to be technical, we only know that the bottom 75% were better off, but considering the 75%er - the median household in the eighth decile - was £6 a week better off, I'm going to assume they weren't the tipping point. Also the 6th decile median stayed flat, rather than rose, but considering both 5th and 7th deciles went up slightly, I'm counting that as rounding until shown otherwise.

<< Home

The Standard's Finance editor is an idiot

OK, further to my rant below, my attention has been drawn to the opening paragraph in a comment piece in the Evening Standard today. It emanates from the Evening Standard's Finance Editor, one Anthony Hilton, who either cannot read a press release or is terminally dishonest.

Mr Hilton says:

"The hollowness of the Government pledge not to increase income tax was revealed yesterday in official figures which showed most people were worse off last year than 12 months previously because pay after taxes had fallen for the first time in 15 years. "

This is absolute rubbish. The IFS data shows clearly that most people (or more technically, Households) were better off in 2003/4. Just to show you how idiotic this paragraph is, here is the second Bullet point of the IFS press release.

"Almost two-thirds of the population lived in households with incomes below the national average, which is pulled up by the incomes of the very well off. The income of the household in the middle of the income distribution – the median – rose by just under £2 a week in 2003/04 to £336 a week. This 0.5% increase was much smaller than those seen in previous years under this government."

So The Evening Standard's Finance Editor thinks that if the median household is better off, "most" people are worse off? In fact it's worse that that. Given the data, we'd assume that around 60% of households were better off in 2003/4*. So the evidence suggests exactly the opposite of what is reported in the Evening Standard today.

My conclusion? Don't trust financial advice given by a man who doesn't understand what average means.

(oh and later on he translates "average income" into "average household", which I'd argue is very misleading, because the average income of me, you and Bill Gates is pretty bloody high.)

*I haven't seen the data, but if 65th percentile Households on average income are £1 worse off, and the median household is £2 better off, then a 60% breakeven point seems a reasonable working assumption until disproved.

<< Home

When journalists attack: number 134 in a never ending series

Today the Institute of Fiscal Studies announced that average incomes fell by 0.2% between 2002/3 and 2003/4. Cue apocalyptic coverage. The Times lead with “Labour defends Tax assault on Middle England”.

The Mail, as impartial and fair minded as ever (no on-line version, to my great regret) asked “has Labour’s bubble burst” and claimed that “swingeing tax increases filtered down to householders, leaving them worse off than they were the year before” and so on.

However, what isn’t mentioned in the coverage is that “Average income” is two thirds of the way up the scale (because it takes a lot of me to balance one Howard Flight). Median Income, or to put it more helpfully, the median schmoe on the Clapham omnibus’s take home pay rose by £2 a week. So more than half of the populace were significantly better off.

It’s not a difficult thing to grasp. Horror of horrors, the top third of the population is a pound a week worse off while the bottom two thirds is better off? This must mean a disaster of unheralded proportions. Yet our beloved press corps seem to find it impossible to grasp the difference between an average family and an average income.

The Times leads this story as bad news for middle England- while somehow neglecting to mention that the Time’s definition of “middle” is actually top third England.

Why is this? Well aside from editorial bias, I suspect it is because Journalists have a completely wrong perception about median income levels. I know I do- I am stunned when I discover I am in the top 20% of household incomes.

Take this test to see if you have a clue about income distribution. If you are on £45,000 in London in a two earner household with two young children, are you a) Slightly above average income distribution, b) Somewhere in the top quarter of income distribution, C) one of the top 10% of household incomes in Britain. No prizes for guessing correctly.

You can play this game for yourself here. Full disclosure: I put myself a full fifteen percent below where I ended up. I’m richer than I thought.

But this distortion was not the only stupid error in the coverage of the IFS report. As well as entirely ignoring the median incomes point, mistakenly leaving readers with the impression that someone on the average earnings is an average family, reports distorted other elements of the IFS study or leave out crucial comments. For example, the Times says

“The IFS said that pensioner poverty fell slightly …...”

The IFS press release says “Pensioner poverty continues to fall dramatically when incomes are measured after housing costs (AHC): it fell by a tenth in the single year 2002/03–2003/04, and has fallen by over a quarter since 1998/99.” It also suggests that the reason Child poverty fell less than they expected was sampling error.

The Times goes on.. “but suggested that Labour would miss one of its key child poverty targets…the IFS said that to meet its target to cut child poverty by a quarter by 2004-05 Labour would have to ensure a further 500,000 children were lifted out of poverty next year.”

Which is a quite different emphasis to the IFS which says: “To meet the government’s short-term target, the number of children in poverty will now have to fall by 300,000 before housing costs and 500,000 after housing costs in 2004/05. The former still looks likely to be achieved, but the latter does not. Having said this, sampling error means that there is always considerable uncertainty around the data in any given year.”

Anyway, my point is simply that in a moderately simple story about tax, incomes and averages, two of our main newspapers left their readers with an extremely distorted view of the data and its import. Do we deserve better reporting or better newspaper owners?

<< Home

Monday, March 28, 2005

Bank Holiday things I’d want to know…

I never knew you could have Easter without a lot of football. It’s a strange lonely experience. Still, the media have decided to turn on the Tories, so atleast someone’s getting a kicking this Easter.

It is a little unfortunate for Michael Howard that his first policy not aimed at attacking gyppos, kosovans and people who look shifty was launched this weekend. It’s actually an intriguing proposal, a £50 a week tax credit that goes towards childcare provided by friends, relatives and so on.

Philosophical inconsistency aside, it’s actually a proposal worth considering, even though it would cost a fortune (paid for how?). I suspect the argument against it is that if you make payments to “informal” childcare providers, you have no idea if the childcare is actually being provided and you open the door to massive fraud. It also perhaps open the doors to somewhat unsavoury people offering “informal childcare”- but enough, we’re on the verge of taking a policy seriously and considering it on the merits and frankly, that would be too horrible. Instead, let’s indulge in some badly informed speculation. Just like proper journalists.

Three things I’d want to know if I was a Westminster journalist:

1. If Howard Flight is deselected for saying the James review is just the beginning of savings that could be made, how can John Redwood still be in the shadow cabinet for saying that the current cuts plan is just a “downpayment” towards further tax cuts?

2. How come Racist joke maven Ann Winterton is a Tory candidate at this election, but Howard Flight is not? Is there a new standard for what sins disqualify someone from being a Conservative candidate?

3. What do MPs who are members of the Conservative way forward group (Member of board, Iain Duncan Smith), think about cutting taxes, Howard flight and the prospect of never again saying what they believe.

Three Things I’d tell Michael Howard if I were a Tory spin doctor:

1. Stop saying “If elected we won’t say one thing and do another, we won’t murder babies and sell them for drugs, we won’t dance naked around a deconsecrated church at midnight”. It’s the accusation we remember, not the denial. As Lyndon Johnson may have said apropos of unfounded porcine intercourse allegations, “Of course it’s not true, I just want to hear him deny it.”

2. Clear your kitchen table of fruit larger than your head before doing live TV interviews.

3. Try smiling. Please. Shrug, emote, laugh, tell a joke... Right now you look like you might just decapitate someone. Relax. You are the victim of a idiot shooting his mouth off, the press pack don’t want to hate yu for this, they want expressions of regret, sorrow, yet firm resolve. Just don’t make them think “something of the night” when you wield the axe.

Three things I’d want to know if I was a Labour spin doctor:

1. The Tories have a poster up on Early Release. Let’s assume the Tories have a heart-rending example of a child attacked by a criminal released on “early release” lined up for the election. How will you cope with that?

2. Other Conservative MP’s and PPCs have recently advocated larger Tax cuts. Do they still hold to that position? If not, why not?

3. How do you turn a tactical opening on Tax cuts that has turned into an easy to understand “Tory chaos” story back into a political story about the impact of cuts in public spending?

Finally the question every political strategist worth their salt should be trying to find the answer to. If dog whistle campaigning works, how many people are considering changing their vote as a result? My own suspicion is that it simply isn’t enough. Who we’re looking for are 2001 Labour supporters who agree with Tougher immigration, asylum and crime polices. Are they shifting? And in what numbers?

<< Home